Retinoids and their receptors directly modulate the transcription of specific genes, whose products influence growth, differentiation, or apoptosis. Alternatively, retinoid-activated transcription factors can stimulate cascades of signaling that indirectly alter a cell behavior. Thus, retinoids directly and indirectly initiate profound developmental changes. The investigators have shown that retinoic acid (RA) and bone morphogenetic proteins (BMPs) 2 and 4, interact to influence embryonic cell behavior. First, RA directly induces the Bmp2 gene. Second, cells exposed to RA or BMP2 or BMP4 alone or the combination of RA and a BMP behave differently. Changes in differentiation, growth rate, and the induction of apoptosis occur. The proposed experiments will augment our understanding of developmental and teratogenic mechanisms involving directly activated genes, including Bmp2. Aim 1. To determine whether or not the Bmp2 retinoic acid response element (RARE) is required for normal Bmp2 expression in mice. This aim will test the hypothesis that an endogenous retinoid signal is necessary for Bmp2 expression by comparing the developmental expression of B-galactosidase reporter genes driven by Bmp2 sequences containing the RARE to those lacking the RARE. Aim 2. To determine whether or not teratogenic levels of retinoids induce aberrant Bmp2 expression. This aim will test the hypothesis that aberrant Bmp2 expression is induced by retinoids by examining the expression of ,B-galactosidase reporter genes driven by Bmp2 sequences in RA-treated embryos. Aim 3. To elucidate the genetic regulatory elements controlling the expression of the Bmp2 gene in F9 embryonal carcinoma cells induced to differentiate into parietal endoderm by RA and increased cAMP levels. This aim will continue our systematic in vitro dissection of the genetic elements controlling Bmp2 expression by identifying the cAMP response element that modulates transcription only in RA-treated cells. Aim 4. To identify genes directly activated by RA by trapping murine RAREs in yeast. This aim will identify genes directly induced by RA and thus primary in controlling signaling cascades leading to changes in cell behavior.